Methods and apparatus having an integrated circuit attached to fused silica

0001This application claims the benefit of U.S. Provisional Application No. 60/802,476, filed May 22, 2006.

BACKGROUND OF THE INVENTION

0002Over the past 20 years, advances in the field of microelectronics have enabled the realization of microelectromechanical systems (MEMS) and corresponding batch fabrication techniques. These developments have allowed the creation of sensors and actuators with micrometer-scale features. With the advent of the above-described capability, heretofore implausible applications for sensors and actuators are now significantly closer to commercial realization.

0003In parallel, much work has been done in the development of pressure sensors. Pressure sensors are disclosed, for example, in U.S. Pat. No. 6,111,520, issued Aug. 29, 2000; U.S. Pat. No. 6,278,379, issued Aug. 21, 2001; U.S. Pat. No. 6,855,115, issued Feb. 15, 2005; U.S. Patent Publication No. 2003/0136417, dated Jul. 4, 2003; U.S. patent application Ser. No. 10/215,377, filed Aug. 7, 2002; U.S. patent application Ser. No. 10/215,379, filed Aug. 7, 2002; U.S. Patent Publication No. 2005/0187482, dated Aug. 25, 2005; U.S. Patent Publication No. 2006/0287602, dated Dec. 21, 2006; U.S. Patent Application Serial No. 2006/0177956 dated Aug. 10, 2006, and U.S. Patent Application Serial No. 2006/0241354, dated Oct. 26, 2006, all of which are incorporated herein by reference.

0004In particular, absolute pressure sensors, in which the pressure external to the sensor is read with respect to an internal pressure reference, are of interest. The internal pressure reference is a volume within the sensor, sealed, which typically contains a number of moles of gas (the number can also be zero, i.e. the pressure reference can be a vacuum, which can be of interest to reduce temperature sensitivity of the pressure reference as known in the art). The external pressure is then read relative to this constant and known internal pressure reference, resulting in measurement of the external absolute pressure. For stability of the pressure reference and assuming the temperature and volume of the reference are invariant or substantially invariant, it is desirable that the number of moles of fluid inside the reference does not change. One method to approach this condition is for the reference volume to be hermetic.

0005The term hermetic is generally defined as meaning “being airtight or impervious to air.” In reality, however, all materials are, to a greater or lesser extent, permeable, and hence specifications must define acceptable levels of hermeticity. An acceptable level of hermeticity is therefore a rate of fluid ingress or egress that changes the pressure in the internal reference volume (a.k.a. pressure chamber) by an amount preferably less than 10 percent of the external pressure being sensed, more preferably less than 5 percent, and most preferably less than 1 percent over the accumulated time over which the measurements will be taken. In many biological applications, an acceptable pressure change in the pressure chamber is on the order of 1.5 mm Hg/year.

0006The pressure reference is typically interfaced with a sensing means that can sense deflections of boundaries of the pressure reference when pressure external to the reference changes. A typical example would be bounding at least one side of the pressure reference with a deflectable diaphragm or plate and measuring deflection of the diaphragm or plate by use of, among other techniques, a piezoresistive or a capacitance measurement. If the deflection of the diaphragm or plate is sufficiently small, the volume change of the pressure reference does not substantially offset the pressure in the pressure reference.

0007These approaches may require an electrical feedthrough to the hermetic environment (e.g., to contact electrodes inside the hermetic pressure reference), for connection to outside electronics to buffer or transmit the signal. Alternatively, electronics may be incorporated within the reference cavity, requiring power to be conducted into the hermetic environment. To maintain stability of the pressure reference, these seals should also be hermetic, resulting in the necessity to develop a feedthrough technology for contacts through the cavity walls. As is known in the art, such feedthrough points are typically sites for failure of hermeticity. This problem is further exacerbated when miniaturizing the sensor, since the total volume of material available for hermetic sealing shrinks proportionally and the reliability of the feedthrough is also greatly reduced. In the limit of ultraminiaturized sensors, such as those producible using microelectromechanical systems (MEMS) technology, one of the major challenges to enabling the use of such devices in applications where they are physically connected to other devices has been the creation of reliable hermetic packaging that provides feedthroughs that enable exchange of power and information with external electronics.

0008Design criteria for ultraminiature packaging that overcomes the aforementioned shortcomings are as follows: The packaging must exhibit long term hermeticity (on the order of the life of the sensor, which in some cases can exceed tens of years). Feedthroughs must be provided through the hermetic package that do not introduce new or unnecessary potential modes of failure. The feedthroughs will constitute a necessary material interface, but all other interfaces can and should be eliminated. In other words, the number and area of material interfaces should be minimized to reduce the potential for breach of hermeticity. The materials selected must be compatible with the processes used to fabricate the package as well as sufficiently robust to resist deleterious corrosion and biocompatible to minimize the body's immune response. Finally, the packaging should be amenable to batch fabrication.

0009In the past, many methods for creating such hermetic packages have been proposed. One approach used in the past to create the pressure cavity is anodic bonding to create a silicon-to-glass seal. A borosilicate glass is required for this method. Another technique utilized in the creation of hermetic packages is eutectic bonding to create a silicon to metal hermetic seal, e.g. Au to Si. Both of these bonding methods used to create the pressure cavity introduce a large area along the perimeter of the material interface of the pressure cavity package which presents opportunity for failure, e.g. through corrosion. These methods for creating the pressure cavity do not minimize the area of the material interface as is desirable. A desirable improvement to the construction of the pressure cavity would minimize the material interface to the hermetic electrical feedthroughs, and, even further, minimize the number and area of material interfaces in those feedthroughs.

0010Previous attempts to create hermetic feedthroughs also fall short of the above-stated requirements. Many prior art hermetic feedthroughs are too large and not amenable to the required miniaturization for pico to nanoliter volume packaging achievable by MEMS or similar approaches. Furthermore, earlier attempts to create feedthroughs in pico to nanoliter packaging are prone to corrosion because of the materials used in construction or are sufficiently complicated that they introduce more material interfaces than are necessary. A representative feedthrough approach, known as a “buried” feedthrough, is illustrated in <figref idref="DRAWINGS">FIGS. 1-5</figref>. One method for creating a buried feedthrough is as follows: a metal <b>10</b> is deposited onto substrate <b>12</b> in a predefined pattern, as shown in <figref idref="DRAWINGS">FIG. 1</figref>. An insulating layer <b>14</b> is deposited on top of the metal layer, as shown in <figref idref="DRAWINGS">FIG. 2</figref>, and this insulating layer <b>14</b> is polished to planarize this surface. In <figref idref="DRAWINGS">FIG. 3</figref> an etchant has been used to expose the metal layer at input and output sites <b>16</b>, <b>18</b> for the feedthroughs. In <figref idref="DRAWINGS">FIG. 4</figref>, another substrate <b>20</b> is bonded on top of this structure, forming a hermetic cavity <b>22</b>. A eutectic bonding method is illustrated, which involves the use of gold deposits <b>24</b> interposed between the insulating layer <b>14</b> and the upper substrate <b>20</b> to bond the upper substrate to the insulating layer. In <figref idref="DRAWINGS">FIG. 5</figref> the upper substrate <b>20</b> is machined to expose the external feedthrough <b>18</b>. An electrical conductor can now be connected to the external feedthrough <b>18</b>, whereupon it is conducted through the metal <b>10</b> to the internal feedthrough <b>16</b> within the hermetically sealed chamber <b>22</b>.

0011This prior art buried feedthrough suffers a number of disadvantages. First, there are numerous material interfaces: an interface <b>30</b> between the lower substrate <b>12</b> and the metal <b>10</b>; an interface <b>32</b> between the metal <b>12</b> and the insulating layer <b>14</b>, an interface <b>34</b> between the insulating layer <b>14</b> and the gold <b>24</b>; and an interface <b>36</b> between the gold <b>24</b> and the upper substrate <b>20</b>, all of which create potential paths for infusion into or effusion out of the hermetic chamber <b>22</b>. The creation of this buried feedthrough also introduces increased processing steps. Further, the insulating layer material is cited as being prone to corrosion in certain environments, e.g. the human body. Corrosion issues may be further exacerbated by the application of electrical bias to metal <b>10</b> which may be required in certain applications. Thus prior art hermetic feedthroughs fall short of meeting the constraints outlined above.

0012Also, many prior art attempts to provide pressure sensors utilize silicon as a substrate material. If the package is implanted in vivo, silicon is not an optimal material choice. Silicon invokes an undesirable immune response over other, more inert materials such as fused silica. If silicon is used, a coating must be applied to ensure biocompatibility. Such a coating increases the package size, thereby decreasing the benefits of miniaturization, and introduces an undesirable additional processing step in the manufacture of the package.

0013Additionally, prior art devices commonly employ the use of borosilicate glass as part of the pressure cavity. The ions in borosilicate glass constitute an impurity in the glass. The barrier to diffusion of water decreases as the purity of glass decreases. This makes use of impure glass undesirable in such applications.

0014Thus a need exists for hermetic pico to nanoliter packaging with electrical feedthroughs for use in biological environments, such packaging being constructed of high-purity materials and having a reduced number and area of material interfaces.

SUMMARY OF THE INVENTION

0015Disclosed are micromachinable, hermetic, pico to nanoliter-volume pressure cavities. Such a pressure cavity uses high-purity materials and provides reliable electrical feedthroughs. The pressure cavity is constructed of a ceramic material and is optionally fused together so that there is no interface of material where two substrates have been joined to create a cavity. Furthermore, feedthroughs establishing electrical communication within said cavity are formed in at least one of the substrates. The feedthroughs themselves are configured in such a way that the number and area of material interfaces is minimized. Such feedthroughs constitute the only site for material interface in the sensor package, thereby decreasing the number of potential leak sites in and increasing the reliability of the hermetic package. Pressure cavities and sensors are manufactured using microelectromechanical systems (MEMS) fabrication techniques, which allow creation of a device that is small, accurate, precise, durable, robust, biocompatible, and insensitive to changes in body chemistry or biology.

0016The pressure cavities further comprise a sensor that can be incorporated into harsh and biological environments. One example of such an environment is a medical lead or catheter implanted, acutely or chronically, into the human body. The sensor is configured to measure one or more physical properties such as pressure or temperature. Communication between the sensor and another device can be established by, e.g., using wires fixed to bonding pads on the exterior of the sensor packaging that are configured so that they are in electrical contact with the hermetic feedthroughs. As another example, the hermetic electrical feedthrough can have a wire extending from the feedthrough, and contact with the pressure cavity can be accomplished via connection with this wire. Devices in electrical communication with sensors may be either implanted or external to the body. Sensors are sufficiently small to allow for incorporation into medical leads or catheters that are twelve French or smaller, preferably six French or smaller, without causing abrupt changes in geometry of the lead or catheter, and require minimal power to perform their intended function.

0017In one embodiment, a wired sensor comprises a hermetic pressure cavity. The pressure cavity further comprises a capacitor configured so that the characteristic capacitance value of the capacitor varies in response to a physical property, or changes in a physical property, of a patient. The electrodes of the capacitor are substantially planar and are arranged substantially parallel to and spaced apart from one another. The pressure cavity has at least one deflectable region in mechanical communication with at least one of the capacitor electrodes. Additionally, electrical feedthroughs are formed through the substrate defining the pressure cavity and allow for the sensor to receive power and signals, and return information to either implanted or extracorporeal external electronics.

0018In another embodiment, a wired sensor comprises a hermetic pressure cavity. The pressure cavity further comprises a Wheatstone bridge configured so that the resistance value of said bridge varies in response to a physical property, or changes in a physical property, of a patient. The pressure cavity has at least one deflectable region in mechanical communication with at least one of the resistors comprising the bridge. Additionally, electrical feedthroughs are formed through the substrate and allow for the sensor to receive power and signals, and return information to external electronics. It is a further aspect that only a portion of the Wheatstone bridge be located within the pressure cavity, the other portion being contained within external electronics.

0019In yet another embodiment, a wired sensor further comprises on-board (i.e., within the sensor package) electronics, e.g., a silicon chip bearing electronics. The variable capacitive or resistive element and the on-board electronics can be maintained in separate cavities in electrical communication with one another by hermetic feedthroughs formed through a middle substrate. Feedthroughs establishing electrical communication with the sensor exterior may be configured so that moisture does not affect the electronics over the life of the sensor and, optionally, are also hermetic. This configuration offers the advantage that the feedthroughs to the on-board electronics act as a redundant barrier to any potential breach of the hermeticity of the pressure cavity. Alternatively, the capacitor and on-board electronics can be contained within a single hermetic cavity. This configuration offers the advantage of decreased manufacturing steps, thereby lowering the overall cost to produce the sensor. In either case, hermetically sealed vias, which are themselves optionally hermetic, formed through the substrates comprising the external walls allow for the sensor to receive power and return information to external electronics.

BRIEF DESCRIPTION OF THE DRAWINGS

0020<figref idref="DRAWINGS">FIG. 1</figref> is a schematic representation of a first step in manufacturing a PRIOR ART hermetic chamber with electrical feedthroughs.

0021<figref idref="DRAWINGS">FIG. 2</figref> is a schematic representation of a second step in manufacturing a PRIOR ART hermetic chamber with electrical feedthroughs.

0022<figref idref="DRAWINGS">FIG. 3</figref> is a schematic representation of a third step in manufacturing a PRIOR ART hermetic chamber with electrical feedthroughs.

0023<figref idref="DRAWINGS">FIG. 4</figref> is a schematic representation of a fourth step in manufacturing a PRIOR ART hermetic chamber with electrical feedthroughs.

0024<figref idref="DRAWINGS">FIG. 5</figref> is a schematic representation of a completed PRIOR ART hermetic chamber with electrical feedthroughs.

0025<figref idref="DRAWINGS">FIG. 6</figref> is a schematic representation of an exemplary hermetic chamber with electrical feedthroughs.

0026<figref idref="DRAWINGS">FIGS. 7-25</figref> are schematic representation of the steps in manufacturing the hermetic chamber of <figref idref="DRAWINGS">FIG. 6</figref>.

0027<figref idref="DRAWINGS">FIG. 26</figref> is a schematic representation of an exemplary hermetic chamber with electrical feedthroughs.

0028<figref idref="DRAWINGS">FIG. 27</figref> is an electrical schematic of a piezoresistive transduction scheme for measuring changes in the position of the deflectable region in the pressure cavity of the hermetic chambers of <figref idref="DRAWINGS">FIGS. 6 and 26</figref>.

0029<figref idref="DRAWINGS">FIG. 28</figref> is a schematic representation of another exemplary hermetic chamber with electrical feedthroughs.

0030<figref idref="DRAWINGS">FIG. 29</figref> is a schematic representation of another exemplary hermetic chamber with electrical feedthroughs.

0031<figref idref="DRAWINGS">FIG. 30</figref> is a schematic representation of yet another exemplary hermetic chamber with electrical feedthroughs.

DETAILED DESCRIPTION OF THE INVENTION

0032Referring now to the drawings, in which like numerals indicate like elements throughout the several views, <figref idref="DRAWINGS">FIG. 6</figref> illustrates a sensor <b>50</b> that includes a pressure cavity body <b>51</b> defining an internal pressure chamber <b>52</b>. One of the walls defining the pressure cavity <b>52</b> comprises a deflectable region <b>54</b> configured to deflect under a physiologically relevant range of pressure. In a preferred embodiment, a wall of the pressure cavity body <b>51</b> is thinned relative to other walls of the pressure cavity body to form the deflectable region <b>54</b>. The sensor <b>50</b> can be fabricated using micro-machining techniques and is small, accurate, precise, durable, robust, biocompatible, and insensitive to changes in body chemistry or biology. Additionally, the sensor <b>50</b> can incorporate radiopaque features to enable fluoroscopic visualization during placement within the body. The sensor <b>50</b> is preferably formed using electrically insulating materials, particularly biocompatible ceramics, as substrate materials. Suitable materials are selected from a group comprising glass, fused silica, sapphire, quartz, or silicon. In one embodiment, fused silica is the substrate material.

0033A capacitor comprises a pair of lower electrodes <b>56</b>, <b>57</b> located on a first wall <b>58</b> of the chamber <b>52</b>. The two lower electrodes <b>56</b>, <b>57</b> are electrically isolated from one another. A third electrode <b>60</b> is disposed on an opposite wall <b>62</b> of the pressure cavity <b>52</b> in parallel, spaced apart relation to the lower electrodes <b>56</b>, <b>57</b>. The upper electrode <b>60</b> is mechanically coupled to the deflectable region <b>54</b>. As ambient pressure increases, the deflectable region <b>54</b> moves inward, displacing the upper electrode <b>60</b> toward the lower electrodes <b>56</b>, <b>57</b>, thereby changing the characteristic capacitance value of the capacitor.

0034The capacitor configuration depicted here is one example where the lower capacitor electrode consists of two electrically isolated regions, <b>56</b> and <b>57</b>, although other configurations are possible and obvious to one skilled in the art.

0035The lower portion of the pressure cavity <b>52</b> comprises passages <b>64</b>, <b>65</b> that traverse the hermetic pressure cavity body <b>51</b> and are in contact with the electrodes <b>56</b>, <b>57</b>. As shown in <figref idref="DRAWINGS">FIG. 6</figref>, electrical contact pads <b>66</b>, <b>67</b> can be formed on the back side of the electrodes <b>56</b>, <b>57</b> and extend to the exterior of the housing, thereby providing a region on the exterior of the sensor <b>50</b> configured with sufficient dimensions so as to allow for a means for connection with external electronics. As an alternative, the passages <b>64</b>, <b>65</b> can be filled with an electrically conductive material, with contact pads <b>66</b>, <b>67</b> in electrical communication with the electrodes <b>56</b>, <b>57</b> by way of the conductive material <b>68</b>. The electrode <b>56</b>, the passage <b>64</b>, and, if present, the electrical contact pad <b>66</b> and any electrically conductive material <b>68</b> filling the passage <b>64</b> comprises a first electrical feedthrough <b>70</b>. The electrode <b>57</b>, the passage <b>65</b>, and, if present, the electrical contact pad <b>67</b> and any electrically conductive material <b>68</b> filling the passage <b>65</b> comprises a second electrical feedthrough <b>71</b>.

0036It is a preferred embodiment that the metal-fused silica interface between the lower electrodes <b>56</b>, <b>57</b> and the interior surface of the pressure cavity body <b>51</b> be hermetic. The electrical contact pads <b>66</b>, <b>67</b> can occupy either all or part of the passages <b>64</b>, <b>65</b>. A variety of metal deposition techniques can be used (e.g., electroplating, use of molten metal, or PVD) depending on the choice of metal and desired material properties. In the case of a partially-filled feedthrough passage <b>64</b>, <b>65</b>, a void inside the feedthrough passages and above the electrical contact pads <b>66</b>, <b>67</b> will remain. In order to fill these voids and to enhance the strength of the feedthroughs <b>70</b>, <b>71</b>, any remaining space in the passages <b>64</b>, <b>65</b> can be filled with a ceramic material. Glass frit is one example of a ceramic material that can be used to fill the remaining space and heated sufficiently that the material flows, thereby eliminating any voids in the ceramic material. In the case of metal-filled feedthrough cavities, the pads <b>66</b>, <b>67</b> on the exterior of the package are formed by, e.g., fusion bonding, low pressure plasma spray, laser welding, electroplating or PVD, depending on the choice of metal and the desired material properties. The electrical contact pads <b>66</b>, <b>67</b> provide a site to connect to external electronics.

0037Suitable non-refractory metals for the electrical feedthroughs include gold, platinum, nickel, and silver and alloys thereof. Suitable refractory metals include niobium, titanium, tungsten, tantalum, molybdenum, chromium, and a platinum/iridium alloy and alloys thereof. If refractory metals are used to construct the feedthroughs, either alternating or direct current may be used to bias the sensors by external electronics. If any other metals are used, the sensors should be biased under AC power to prevent the onset of bias-induced corrosion.

0038The pressure cavity <b>52</b> is hermetic for the following reasons. First, the pressure cavity body <b>51</b> is formed of a hermetic material and is a unitary structure, meaning there are no seams or bi-material joints that can form a potential path for gas or fluid intrusion into the pressure chamber other than the passages <b>64</b>, <b>65</b>, which themselves are hermetically sealed. One reason for the hermeticity of the passages <b>64</b>, <b>65</b> is that the electrodes <b>56</b>, <b>57</b> are hermetically imposed onto the wall <b>58</b> over the feedthroughs. The electrodes <b>56</b>, <b>57</b> (along with any other metallic structure fixed to the ceramic substrate) optionally form an intermetallic compound. An intermetallic compound is formed between a metal and a substrate when chemical reactions take place that result in the formation of covalent bonds between two or more elements, with at least one of the elements coming from the substrate and one from the metal. Optionally, the material <b>68</b> filling the passages <b>64</b>, <b>65</b> is itself capable of hermetic sealing such that the interface between the material <b>68</b> and the material defining the feedthrough passages is also hermetic. Thus gas or fluid would have to pass through or around the material <b>68</b> in the passages <b>64</b>, <b>65</b> and pass through or around the electrodes <b>56</b>, <b>57</b> before it could enter the pressure chamber and compromise its integrity. And finally, the passages <b>64</b>, <b>65</b> are small, thereby minimizing the area of interface and reducing the probability of flaw creation and propagation. In the disclosed embodiments, the passages have cross-sectional areas ranging from 10⁻⁶to 10⁻⁹square meters.

0039A method of fabricating the sensor <b>50</b> depicted in <figref idref="DRAWINGS">FIG. 6</figref> is based on the micromachining of two substrates that are subsequently brought into contact and cut into individual sensors. The manufacturing process described herein and illustrated in <figref idref="DRAWINGS">FIGS. 7-25</figref> comprises a series of etching, deposition and patterning processes to create depressions and electrodes on the surfaces of the substrates. More specifically, a first substrate is subjected to a series of processes to create local depressions of known depth and to deposit and pattern thin film electrode(s) at the bottom of the depressions. Next, a second substrate is subjected to similar processing as the first substrate to create complementing electrode(s) whose overall footprint and in-plane position correspond to the footprint and in-plane position(s) of the electrode(s) on the first substrate. Creation of depressions in the surface of the second substrate is optional and depends on the desired final configuration of the sensor. The first substrate is then subjected to additional processing on the side of the substrate opposite the previously formed electrode(s) to physically remove material through the entire thickness of the substrate to create the passages that are the first step in creating electrically conductive feedthroughs that allow for electrical communication with the hermetic cavity. The configuration of the electrodes and the passages can be altered to provide for a variety of configurations, such modifications providing manufacturing and/or performance advantages. The two substrates are then brought into intimate contact with the electrodes facing one another. The substrates form a temporary bond due to the presence of Van der Waals forces. The electrodes on opposing substrates are separated by a gap of known value, i.e., the difference between the sum of the depths of the recessed region and the sum of the thicknesses of the electrodes. A laser is then used to excise the sensor into its final overall dimensions from the two-substrate stack.

0040The laser cutting operation fuses the substrates, hermetically sealing the sensor and trapping air or any other desirable gas in the hermetic cavity of the sensor, or creating a vacuum within the hermetic cavity of the sensor. In one example, a CO₂laser operating at a peak wavelength of ten microns is used to hermetically seal and to reduce the sensor to its final size. The laser energy is confined to a precise heat effect zone where the substrates are fused, eliminating any material interface between the original substrates.

0041The resulting hermetic package presents electrical feedthroughs <b>70</b>, <b>71</b> created in the sensor body <b>51</b> that allow for communication between components inside the hermetically-sealed sensor <b>50</b> and external electrical components. The feedthroughs <b>70</b>, <b>71</b> are small, thereby minimizing the area of interface. Such feedthroughs interface with the substrate at areas ranging from 10⁻⁶to 10⁻⁹square meters.

0042For the purpose of illustration, sensors according to <figref idref="DRAWINGS">FIG. 6</figref> have been manufactured that displayed 0.1-10 picofarads capacitance and, more particularly, 1-5 picofarads capacitance. Also, sensitivities of the device easily can be, e.g., 0.1 KHz/mmHg.

0043The manufacturing of the sensor <b>50</b> depicted in <figref idref="DRAWINGS">FIG. 6</figref> from the substrate (a.k.a. wafer) level to the final device is described in greater detail below. For clarity, the manufacture of the sensor <b>50</b> is described on a single-sensor basis, although it will be understood that multiple sensors can be created simultaneously on the substrate in a batch process to increase manufacturing efficiency.

0044The lower substrate is processed to create a recessed region in its surface and thin film electrodes at the bottom surface of each recessed region. Creation of a recessed region with known geometry comprises the steps of (i) depositing and patterning a mask at the surface of the wafer, (ii) etching the wafer material through openings in the mask, and (iii) removal of the mask.

0045One method for creating the desired recessed region is depicted in <figref idref="DRAWINGS">FIGS. 7-20</figref> and described as follows. Referring first to <figref idref="DRAWINGS">FIG. 7</figref>, a thin metallic film <b>100</b> is deposited at the surface of a fused silica substrate <b>102</b> using a physical vapor deposition system (e.g., an electron-beam evaporator, filament evaporator, or plasma assisted sputterer). This thin film layer <b>100</b> will form a mask used to create a recessed region in the upper surface of the substrate <b>102</b>. The nature and thickness of the metal layer <b>100</b> are chosen so that the mask is not altered or destroyed by a glass etchant. For the purpose of illustration, Cr/Au or Cr/Ni are examples of suitable mask materials. A representative Cr/Au mask is 100-200 Angstroms of chromium and 1000-3000 Angstroms of gold.

0046As can be seen in <figref idref="DRAWINGS">FIG. 8</figref>, a layer <b>104</b> of photoresist is formed atop the thin metal film <b>100</b> and substrate <b>102</b>. Then, as shown in <figref idref="DRAWINGS">FIG. 9</figref>, a mask <b>106</b> having a rectangular opening is positioned over the photoresist layer <b>104</b>, and ultraviolet light, indicated by the arrows <b>107</b>, is directed through the mask <b>106</b> onto the exposed portions of the photoresist layer <b>104</b>. The exposed photoresist defining the body of the rectangular region is removed via the appropriate etchants, as illustrated in <figref idref="DRAWINGS">FIG. 10</figref>.

0047Referring now to <figref idref="DRAWINGS">FIG. 11</figref>, etchants are used to etch away the rectangular portion of the thin metallic film <b>100</b> exposed through the patterned photoresist layer <b>104</b>. When the remaining photoresist material is removed, such as by using an appropriate organic solvent, the substrate <b>102</b> is left with a metallic mask <b>108</b> defining a rectangle <b>110</b>, as illustrated in <figref idref="DRAWINGS">FIG. 12</figref>.

0048A glass etchant is used to etch the portion of the upper surface of the substrate <b>102</b> that is exposed through the mask <b>108</b>. To accomplish this, the substrate <b>102</b> is placed in a fixture that prevents the etchant from contacting the un-masked back side of the substrate and is then submerged in a solution containing hydro-fluoric acid, resulting in etching of the masked substrate only where the fused silica is exposed. The substrate <b>102</b> is removed from the acid when the substrate has been etched to the desired depth, usually on the order of 1-3 micrometers. The resulting etched substrate <b>112</b> with rectangular recessed region <b>114</b> is shown in <figref idref="DRAWINGS">FIG. 13</figref>. Then, as shown in <figref idref="DRAWINGS">FIG. 14</figref>, the mask <b>108</b> is removed from the etched substrate <b>112</b> using selective etchants and solvents.

0049The etched substrate <b>112</b> is now primed for creation of electrodes at the bottom of the recessed region <b>114</b>. As shown in <figref idref="DRAWINGS">FIG. 15</figref>, a thin film metal layer <b>120</b> is deposited onto the upper surface of the etched substrate <b>112</b>. For the purposes of illustration, this thin film metal layer <b>120</b> can be composed of elemental chromium and gold. A representative Cr/Au layer is a 100-200 Angstrom seed layer of chromium and 1000-3000 Angstroms of gold. The thin film layer <b>120</b> can also utilize a Ti seed layer and either a Ni or Pt secondary layer. The thickness of this layer is carefully controlled so that, in this embodiment, the metal layer <b>120</b> does not protrude above the level of the original surface of the patterned side of the substrate.

0050Referring now to <figref idref="DRAWINGS">FIG. 16</figref>, a layer of photoresist <b>122</b> is deposited over the surface of the metal layer <b>120</b>. A mask <b>124</b> is positioned over the photoresist layer <b>122</b>, and ultraviolet light, indicated by the arrows <b>125</b>, is directed onto the exposed portions of the photoresist layer, as shown in <figref idref="DRAWINGS">FIG. 17</figref>. Then, as illustrated in <figref idref="DRAWINGS">FIG. 18</figref>, the exposed photoresist is removed, leaving a mask <b>126</b> of photoresist material formed on the upper surface of the metal layer <b>120</b>.

0051Next, the portions of the metal layer <b>120</b> exposed through the mask <b>126</b> are etched away, as illustrated in <figref idref="DRAWINGS">FIG. 19</figref>. In this instance, the patterns defined by the remaining photoresist <b>126</b> represent two side-by-side rectangles whose in-plane, overall foot print is smaller than that of the recessed region <b>114</b>. The rectangles are a few micrometers to tens of micrometers apart and maintain at least a few micrometers wide border separating the rectangles from the perimeter of the rectangular trench <b>114</b>. Subsequently, the photoresist mask <b>126</b> is removed with appropriate organic solvents.

0052As depicted in <figref idref="DRAWINGS">FIG. 20</figref>, the etched lower substrate <b>112</b> is patterned with a rectangular trench <b>114</b> etched into its upper surface, and the base of the rectangular trench contains side-by-side, spaced apart metal electrodes <b>56</b>, <b>57</b> of known thickness. The difference between the height of the upper surface of either electrode, H₁, and depth D₁of the trench <b>114</b> created in the lower substrate <b>102</b>, is substantially constant (excepting for inherent variations in the substrate and patterned metal), and these dimensions are known with great precision, i.e. fractions of micrometers.

0053An optional step involving creation of an intermetallic compound can be performed, e.g., at this step and serves to increase the hermeticity of the metal-substrate interface. An intermetallic compound is created by annealing a metal deposited onto a ceramic substrate at a temperature sufficient to initiate covalent bonding across the substrates. It may be necessary to protect the surface of the metal from oxidation by providing a protective layer to the exposed metal or by performing the annealing step in an inert environment (e.g., vacuum, N₂). One example of an intermetallic compound is a Ti—O—Si system, where titanium is deposited onto a SiO₂substrate. Exposed Ti surfaces are protected from oxidation by a layer of silicon nitride. The metal and underlying ceramic substrate are heated at a ramp rate of, e.g., 4-10 degrees C./minute to between about 700 and about 1100 degrees C. in order to drive the fusion reaction. The temperature is gradually increased and decreased in order to obviate any potential problems with CTE mismatch between the metal and the substrate. If necessary, the protective layer is then removed. In this Ti—O—Si system, either the Ti dissolves significant amounts of oxygen prior to oxide formation enabling the oxygen to react with Si diffusing to the interface, or the stable oxide evolves from TiO to SiO₂in the presence of the Ti-rich phases. Other configurations of metals and substrates can be used to achieve the same effect, e.g., W—Si—O, Mo—Si—O, Ta—Si—O, and Ti—Si—N. To carry out this annealing step, one skilled in the art need only reference the ternary phase diagram to determine sufficient annealing temperatures and to discern the relevant properties of the intermetallic compound.

0054Referring now to <figref idref="DRAWINGS">FIG. 21</figref>, an upper substrate <b>150</b> is micromachined using the same sequence of steps described above to create a rectangular trench <b>152</b> in the fused silica, and the electrode <b>60</b> is created using the same photolithographic process as those described for the lower substrate <b>102</b>. The only change to the preparation of the upper substrate is in the pattern transferred to the second layer of photoresist, i.e. the photoresist layer that serves as a mask for creating the metal electrode. On this substrate <b>150</b>, one continuous rectangle is patterned that maintains a border at least one micrometer thick separating the electrode <b>60</b> from the perimeter of the rectangular trench <b>152</b>.

0055As an optional preparatory step for the upper substrate <b>150</b>, a blanket etch can be performed on the back side using hydrofluoric acid or any other suitable etchant to form the recess <b>54</b> such that overall thickness of the substrate <b>150</b> is reduced to a known thickness that lies in the range of 30-100 micrometers. This step serves to increase sensitivity of the deflectable region of the pressure cavity body <b>51</b> (<figref idref="DRAWINGS">FIG. 6</figref>). Alternatively, the upper substrate <b>150</b> can have an initial thickness in this range, which obviates the need for the above-described step.

0056The substrates <b>112</b>, <b>150</b> are then aligned, subjected to bonding, and reduced to final overall dimension of the sensor as shown in <figref idref="DRAWINGS">FIG. 6</figref> in the following manner: Both the upper and lower substrates <b>112</b>, <b>150</b> are prepared for assembly, e.g., by cleaning. The patterned surfaces of the substrates are faced and aligned so that the corresponding rectangular trenches <b>114</b>, <b>152</b> created in each substrate are positioned directly on top of one another. The two substrates <b>112</b>, <b>150</b> are brought together and placed in intimate physical contact, as shown in <figref idref="DRAWINGS">FIG. 22</figref>. A temporary bond is formed because of Van der Waals forces existing between the two substrates. As previously described, a gap is maintained between the electrodes <b>56</b>, <b>57</b> and the electrode <b>60</b> where the distance between the electrodes is precisely known. Referring to <figref idref="DRAWINGS">FIG. 23</figref>, using a CO₂laser, indicated by the arrows <b>160</b>, the sensor is reduced to its final dimensions. The laser cutting process also seamlessly fuses the upper and lower substrates <b>112</b>, <b>150</b>. The result of the above steps is depicted in <figref idref="DRAWINGS">FIG. 24</figref>. Thus, the rectangular electrodes created combine to form a complete device that displays the electrical attributes of a parallel plate capacitor.

0057With further reference to <figref idref="DRAWINGS">FIG. 24</figref>, the power of the CO₂laser is controlled such that heat damage to the internal components is avoided. Consequently it is possible that some vestige of a seam <b>162</b> may remain between the upper and lower substrates <b>112</b>, <b>150</b>. So long as the outer periphery of the pressure cavity body <b>51</b> is completely fused, the interior chamber <b>52</b> will be hermetic.

0058At some point, the feedthrough passages <b>64</b>, <b>65</b> are created by removing material on the lower surface of the pressure cavity body <b>51</b> to expose the back side of the capacitor electrodes <b>56</b>, <b>57</b>, establishing electrical communication through this location as pictured in <figref idref="DRAWINGS">FIG. 25</figref>. This process step can take place after completion of the electrodes <b>56</b>, <b>57</b> on a single substrate <b>112</b>, after the two substrates <b>112</b>, <b>150</b> have been temporarily bonded, or after the sensors <b>50</b> have been individualized, depending on manufacturing considerations. Either laser ablation or chemical etching or a combination of the two is performed to remove the glass substrate and to expose a portion of the back side of each of the electrodes <b>56</b>, <b>57</b> located on the lower surface of the pressure cavity <b>51</b>. In order to provide for electrical contact pads, any number of techniques can be used to deposit a layer of metal into the passages <b>64</b>, <b>65</b>. The metal choice and deposition technique cannot be chosen independently from one another, but these combinations, along with their respective advantages and shortcoming, are well-known in the art. For purposes of illustration, techniques such as low-pressure plasma spray, electroplating, or screen printing can be utilized to this end. Optionally, if compatible with the deposition technique chosen and the strength of the exposed electrodes, the metal deposition is performed under vacuum. If the feedthrough passages <b>64</b>, <b>65</b> are only partially filled with the electrical contact pad, a ceramic material (e.g., glass frit) can be used to fill the remainder. This would provide mechanical reinforcement to the feedthrough structure.

0059It is a further aspect to provide for a hermetic sensor that incorporates a pressure cavity and additional electrical components that incorporate the above described advantages, with additional functionality and advantages being provided. An exemplary sensor, along with desirable modifications, is depicted in <figref idref="DRAWINGS">FIG. 26</figref> and is further described below.

0060<figref idref="DRAWINGS">FIG. 26</figref> shows a sensor <b>200</b> comprising a sensor body <b>202</b> of fused silica or other suitable material, as discussed above. The sensor body <b>202</b> comprises a lower wall <b>204</b>, an upper wall <b>206</b>, and an intermediate wall <b>208</b>. The intermediate wall <b>208</b> divides the hollow interior of the sensor body <b>202</b> into a lower hermetic chamber (a.k.a. pressure chamber) <b>210</b> and an upper chamber <b>212</b>. A first electrode <b>214</b> is affixed within the lower hermetic chamber <b>210</b> to the lower sensor body wall <b>204</b>. A second electrode <b>216</b> is affixed within the lower hermetic chamber <b>210</b> to the intermediate wall <b>208</b>. A third electrode <b>217</b> is behind and in-plane with the second electrode <b>216</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 26</figref>. The first electrode <b>214</b> is thus arranged in parallel, spaced-apart relation with respect to the second and third electrodes <b>216</b>, <b>217</b> so as to form a gap capacitor. A recess is formed in the lower sensor body wall <b>204</b>, or the substrate comprising the lower sensor body wall is configured to be sufficiently thin, to form a region <b>220</b> that will deflect in response to pressure changes. Because the first electrode <b>214</b> is coupled to the deflectable region <b>220</b>, the distance between the first electrode <b>214</b> changes with respect to the second and third electrodes <b>216</b>, <b>217</b> with variations in external pressure. Thus the characteristic capacitance of a capacitor comprising the first, second, and third electrodes <b>214</b>, <b>216</b>, <b>217</b> changes with movement of the deflectable region <b>220</b>.

0061Also mounted to the intermediate wall <b>208</b> within the lower hermetic chamber <b>210</b> is a fourth electrode <b>224</b>. A fifth electrode <b>226</b> is located on the intermediate wall <b>208</b> within the upper chamber <b>212</b>, which is, optionally, hermetic. A sixth electrode <b>225</b> is behind and in-plane with the fifth electrode <b>226</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 26</figref>. The fourth electrode <b>224</b> is disposed in parallel, spaced apart relation with respect to the fifth and sixth electrodes <b>225</b>, <b>226</b>, separated by the thickness of the intermediate wall <b>208</b>. Because the distance between the fourth electrode <b>224</b> and the fifth and sixth electrodes <b>225</b>, <b>226</b> remains constant, a capacitive circuit comprising the fourth, fifth, and sixth electrodes provides a fixed reference. In the capacitor configuration described above, an example where the need for feedthroughs into the lower hermetic chamber <b>210</b> is eliminated, a capacitor configuration (i.e., a configuration that is physically two capacitors in parallel) that sacrifices capacitance value for ease of manufacture is utilized. Alternative configurations can be provided for, require either one or two feedthroughs into the lower hermetic chamber and are obvious to one skilled in the art.

0062Electrical contact pads <b>230</b>, <b>231</b> are formed on the intermediate wall <b>208</b> within the upper hermetic chamber. A first pad <b>230</b> is located opposite a portion of the second electrode <b>216</b>. A second pad <b>231</b> is located opposite a portion of the third electrode <b>217</b> and is behind and in plane with the first pad <b>230</b> and thus not visible in <figref idref="DRAWINGS">FIG. 26</figref>. A first feedthrough passage <b>236</b> places the first pad <b>230</b> and the second electrode <b>216</b> in communication through the intermediate wall <b>208</b>. A second feedthrough passage <b>237</b> places the second pad <b>231</b> and the third electrode <b>217</b> in communication through the intermediate wall <b>208</b>. The electrical feedthrough passages <b>236</b>, <b>237</b> are filled with a conductive material, such as metal. The second and third electrodes <b>216</b>, <b>217</b> are hermetically imposed against the openings of the passages <b>236</b>, <b>237</b>. Optionally, the pads <b>230</b>, <b>231</b> and the medium filling the passages <b>236</b>, <b>237</b> are hermetic and are hermetically imposed against the openings of the passages <b>236</b>, <b>237</b>. At a minimum, this hermetic imposition of electrodes <b>216</b> and <b>217</b> renders the feedthroughs hermetic. Optionally, electrical contact pads <b>230</b>, <b>231</b> and/or the material filling the feedthrough passages <b>236</b>, <b>237</b> further renders the feedthroughs hermetic.

0063To provide electrical access to the interior of the sensor, fifth and sixth feedthrough passages <b>240</b>, <b>241</b> are provided. The passage <b>240</b> extends from the exterior of the sensor body to the upper chamber <b>212</b>. The passage <b>241</b> also extends from the exterior of the sensor body to the upper chamber <b>212</b> but is behind and in plane with the electrical feedthrough <b>240</b> and thus not visible in <figref idref="DRAWINGS">FIG. 26</figref>. An electrical contact pad <b>242</b> is located within the upper chamber <b>212</b> on the intermediate wall <b>208</b> and is imposed over the passage <b>240</b>. Likewise, an electrical contact pad <b>243</b> is located within the upper chamber <b>212</b> on the intermediate wall <b>208</b> and is imposed over the passage <b>241</b>. The electrical contact pad <b>243</b> is behind and in plane with the electrical contact pad <b>242</b> and is therefore not visible in <figref idref="DRAWINGS">FIG. 26</figref>. Electrical contact pads <b>242</b>, <b>243</b> can be configured to provide a hermetic interface with the intermediate wall <b>208</b>. In the embodiment of <figref idref="DRAWINGS">FIG. 26</figref>, the passages <b>240</b>, <b>241</b> are partially filled with a conductive material such as gold, and electrical connection can be made on the exterior of the sensor body <b>202</b> as described in previous examples. Any remaining voids in the passages <b>240</b>, <b>241</b> are filled with a material <b>248</b> such as glass frit, which fills the space not occupied by the conductive material and enhances the mechanical stability of the feedthrough structure. Optionally, hermetic imposition of the conductive material into the passages <b>240</b>, <b>241</b> further renders the feedthroughs hermetic.

0064The upper chamber <b>212</b> contains one or more electrical components such as a silicon chip <b>250</b> bearing electronics that can act to buffer, to linearize, or otherwise to manipulate the electronic signal from the transducer. The silicon chip <b>250</b> is placed in electrical communication with the electrodes and with an external source by way of the conductive pads <b>230</b>, <b>231</b>, <b>242</b>, <b>243</b>. In one embodiment (not shown), the electronics comprises an A/D converter placed in series with an additional silicon chip bearing electronics. In this case, an additional set of electrical contact pads are provided that allow electrical communication between the A/D converter and the additional electronics.

0065The fabrication of the sensor depicted in <figref idref="DRAWINGS">FIG. 26</figref> is based on the micromachining of three substrates that are subsequently brought into contact and cut into individual sensors. The fabrication of the individual substrates as well as their final assembly is described as follows: The thin metal electrodes <b>216</b>, <b>217</b> having overall, in-plane dimensions of 500 micrometers width, 3-4 mm length and 500 nm thickness, are formed within a recessed region of the same dimensions of the electrode that was previously etched into the surface of a first substrate using photolithography and chemical etching as described for previous examples. The metal electrodes are shorter than the depth of the recessed region by 200 nm. A second substrate has a second recessed region formed therein having a depth of 700 nm and the same cross-sectional dimensions as the recessed region in the upper wafer. A thin metal electrode <b>214</b>, having a thickness of 500 nm and the same overall, in-plane dimensions as electrodes <b>216</b>, <b>217</b>, is then formed into this recessed region. The electrode <b>214</b> is thinner than the depth of the recessed region by 200 nm. When the first and second substrates are bonded together with their respective recessed regions facing each other, a gap of 400 nm is thereby formed between the electrode <b>214</b> and the electrodes <b>216</b>, <b>217</b>. Feedthrough passages <b>236</b>, <b>237</b> are then created from the top surface of the second substrate down to the upper electrodes <b>216</b> and <b>217</b>, using laser rastering and HF etching. Also, electrode <b>224</b> and electrodes <b>225</b>, <b>226</b> are formed on opposite sides of the wall <b>208</b>.

0066Conductive pads <b>230</b>, <b>231</b>, <b>242</b>, <b>243</b> on the top surface of the second substrate can be formed during the feedthrough fabrication sequence. The silicon chip <b>250</b> is then connected to the conductive pads <b>230</b>, <b>231</b>, <b>242</b>, <b>243</b> that were formed during the feedthrough fabrication sequence. A third substrate that has a recess sufficiently deep to contain the silicon chip <b>250</b> and to make contact to the second substrate is added to the assembly. A laser is then used to remove material around the sensor periphery to reduce the sensor to final dimensions. In the disclosed embodiment, the sensor is 750 micrometers wide by 4-5 mm long and 0.6 mm tall. Passages <b>240</b>, <b>241</b> are then created to allow for conductive communication with external electronics.

0067<figref idref="DRAWINGS">FIG. 29</figref> shows a sensor <b>500</b> comprising a sensor body <b>505</b> of fused silica, or other suitable material, as discussed above. The sensor body <b>505</b> comprises a lower wall <b>510</b> and an upper wall <b>515</b>. The lower wall <b>510</b> further comprises a first trench <b>540</b> and a second trench <b>545</b> formed within a portion of the first trench <b>540</b>. The sensor body <b>505</b> further defines a hermetic chamber <b>520</b>. Located within the hermetic chamber on the lower wall <b>510</b> in the area comprising the first trench <b>540</b> are a first electrode <b>525</b> and a second electrode <b>530</b>. The second electrode <b>530</b> is behind and in-plane with the first electrode <b>525</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 29</figref>. A third electrode <b>535</b> is located within the hermetic chamber <b>520</b> on an upper wall <b>515</b> and is positioned such that it is in parallel, spaced-apart relation with respect to the first and second electrodes <b>525</b>, <b>530</b>. The first, second and third electrodes, <b>525</b>, <b>530</b><b>535</b> combine to form a gap capacitor. The second trench <b>545</b> in the lower wall <b>510</b> contains a silicon chip bearing electronics <b>550</b>. Located on the opposite side of the first trench <b>540</b> from the electrodes <b>525</b>, <b>530</b> are a fourth electrode <b>555</b> and a fifth electrode <b>560</b>. The fifth electrode <b>560</b> is behind and in-plane with the fourth electrode <b>555</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 29</figref>. A first wirebond <b>565</b> is established between a site for electrical connection on chip <b>550</b> and first electrode <b>525</b>. A second wirebond <b>570</b> is established between a second site for electrical connection on the chip <b>550</b> and the second electrode <b>530</b>. The second wirebond <b>570</b> is behind and in-plane with the first wirebond <b>565</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 29</figref>. A third wirebond <b>575</b> is established between a site for electrical connection on the chip <b>550</b> and fourth electrode <b>555</b>. A fourth wirebond <b>580</b> is established between a fourth site for electrical connection on the chip <b>550</b> and fifth electrode <b>560</b>. The fourth wirebond <b>570</b> is behind and in-plane with the first wirebond <b>565</b> and is thus not visible in <figref idref="DRAWINGS">FIG. 29</figref>. A third trench <b>585</b> is provided in the upper wall <b>515</b>. The upper wall <b>515</b> of the sensor <b>500</b> acts as a deflective region <b>625</b> and is configured to be sufficiently thin to deflect in response to pressure changes. As explained in previous examples, the third electrode <b>535</b> is coupled to this deflective region <b>625</b> so that the distance between the third electrode <b>535</b> and the first and second electrodes <b>525</b>, <b>530</b> changes with variations in external pressure, thereby changing the characteristic capacitance of the capacitor.

0068Fabrication of the sensor <b>500</b> depicted in <figref idref="DRAWINGS">FIG. 29</figref> involves micromachining of two substrates that are subsequently brought into contact and cut into individual sensors. The fabrication of the individual sensors as well as their final assembly is described as follows: A lower substrate is provided. Into this lower substrate, a first trench is etched that may be 2 mm by 5 mm and 3 micrometers deep using conventional masking processes and wet etching techniques as described in earlier examples. A second trench is then etched in a portion of the lower substrate using the same methods as used in creation of the first trench. The second trench may be 100 micrometers deep, 1.3 mm long, and 0.9 mm wide. Next, the first, second, fourth and fifth electrodes are formed by conventional masking techniques and thin film techniques as described in prior examples.

0069The chip is inserted in the recessed cavity and fixed by eutectic bonding via additional metallic interface such as a soldered preform. Soldered performs can be used to attach a silicon IC to a fused silica substrate. At the wafer level (i.e., before individual chips are diced), a conventional PVD process is used to deposit a Cr/Au film (200-300 Angstroms/3000-5000 Angstroms). The ICs are then individualized from the wafer. Next, the fused silica substrate is modified. A conventional PVD process is used to deposit a Cr/Ni/Au film (200-300 Angstroms/5 micrometers/3000-5000 Angstroms) in a predefined pattern at the intended IC attachment site, in this case, at the base of the recessed cavity in the lower substrate. An Au/Sn solder preform. e.g., a 90% Au and 10% Sn solder preform, is placed onto the Cr/Ni/Au film on the base of the recessed region in the lower substrate. The IC is then positioned on top of the soldered preform such that the Cr/Au film interfaces with the preform. The assembly is then heated on, e.g., a hot plate or rapid thermal processor to a temperature between about 100 and about 400 degrees C. and held at that temperature for between about 10 seconds and 5 minutes. This results in fixation of the IC on the fused silica substrate. Modifications can be made to this process to achieve substantially the same results. For example, the Cr/Ni/Au layer can be placed on the IC and the Cr/Au layer placed on the fused silica. Only one Ni layer is needed in the assembly to achieve the desired result. Also, the surfaces on which PVD is used to deposit metal must be sufficiently clean to achieve adequate adhesion. Use of a solder preform (metallic interface) circumvents the possibility of sensor drift due to outgassing that is possible if a polymer adhesive is used to attach components within a sealed hermetic cavity.

0070After the chip is placed and fixed to the lower substrate, the first, second, third and fourth wire bonds are made to the first, second, fourth and fifth electrodes, respectively, via conventional techniques, e.g., with wire that is 25 micrometers in diameter, for example.

0071Next, the upper substrate is prepared. To this end, a third trench is etched into this wafer using the same techniques as used to create the first and second trenches of the lower substrate. This trench is on the order of 2.5 mm long, 1.7 mm wide, and 0.1 mm deep. The third electrode is then created using the same techniques referenced in the creation of the electrodes on the lower substrate. This third electrode is on the order of 1.4 mm by 2 mm and is made of 500 nm layer of chrome/gold.

0072Subsequent to individual fabrication of the substrates, the upper wafer is oriented with respect to the lower wafer such that the components are aligned as shown in <figref idref="DRAWINGS">FIG. 29</figref>. The wafers form a temporary bond and are, optionally, further subjected to a 200 degrees C. oven for approximately 2 hours in order to increase the bond strength.

0073Passages <b>590</b>, <b>595</b> are created through the exterior surface of the lower substrate using a CO₂laser as described previously. After laser ablation, the passages are optionally subjected to a wet etch. The passages resulting from this process are approximately 200 micrometers at the exterior surface and about 50 micrometers at the interior surface and the back side of the fourth and fifth electrodes are exposed. Next, a metal layer is deposited through a shadow mask using a DC sputterer. The metal used is titanium and the resulting layer is 10 micrometers thick. This titanium layer is intended to establish electrical communication between the interior of the hermetic chamber and the ambient. Furthermore, the titanium layer increases the integrity of the hermetic cavity and provides further corrosion resistance.

0074The capacitor contained within the individualized sensor made with the above geometry is approximately 5 picofarads. It should be apparent to one skilled in the art in light of the above disclosure that modifying the spacing between the electrodes as well as the area of the interposed electrodes increases or decreases the capacitance value.

0075The sensor is then individualized from the two wafer stack using the same method as disclosed in previous embodiments.

0076In summary, improvements disclosed herein include apparatus comprising a glass substrate, a metallic layer disposed on the substrate, and an integrated circuit eutectically bonded to the glass substrate via the metallic layer. The metallic layer may comprise a solder preform or one or more layers of metal deposited on the substrate. In a fully fabricated sensor, the glass substrate comprises one or more trenches. In addition, one or more first electrodes are disposed on a surface of the one or more trenches. Wire bonds are connected between the integrated circuit and the one or more first electrodes. A second glass substrate comprises one or more second electrodes that cooperate with the one or more first electrodes to form a sensor. One or more electrical feedthroughs are disposed through the glass substrate that are coupled to selected ones of the one or more electrodes. Wire bonds are connected between selected pads of the integrated circuit and the one or more electrical feedthroughs. Peripheries of the respective substrates are fused together to form a hermetic sensor.

0077An exemplary method eutectically bonds an integrated circuit to a glass (such as fused silica) substrate. The integrated circuit may be bonded to the substrate using a solder preform. This may be achieved by deposing a solder preform on the substrate and heating the substrate, solder preform and integrated circuit to form a eutectic bond therebetween. This may also be achieved by depositing one or more layers of metal on the substrate to form an alloy, and heating the substrate, one or more layers of metal and integrated circuit to form a eutectic bond therebetween.

0078Processing may also be performed to provide for a hermetic sensor. This is achieved by forming one or more trenches in the glass substrate. One or more first electrodes are then formed on the one or more trenches. Wire bonds are connected between the integrated circuit and the one or more first electrodes. One or more second electrodes are formed on a second glass substrate that cooperate with the one or more first electrodes to form a sensor. Electrical feedthroughs are formed through the glass substrate. Wire bonds are connected between the integrated circuit and the electrical feedthroughs. Peripheries of the substrates are fused together to form the hermetic sensor.

0079Regarding the manufacture of the sensor, the overall size of the resultant device can be reduced through use of an anisotropic etching method (e.g., ICP glass etching, ultrasonic glass etching) instead of isotropic wet etching. If the chip utilized in the example above is 800 by 1300 micrometers, the second trench in the lower wafer can very well be merely 810 by 1310 micrometers. Also, the capacitor area can be reduced to 700 microns by 800 microns by reducing the gap between the electrodes. Furthermore, the thickness of the fused silica package can also be reduced to about 100 micrometers by reducing the thickness of the wall surrounding the hermetic cavity. Thus, it follows that the sensor can be reduced to final overall dimensions of 1 mm by 2.3 mm by 0.6 mm versus the disclosed example that results in a device that is 2 mm by 5 mm by 0.6 mm as shown in <figref idref="DRAWINGS">FIG. 30</figref>. In addition, if no chip is included in the sensor package the sensor (such as that disclosed in <figref idref="DRAWINGS">FIG. 1</figref>) can achieve even smaller geometries. Also, as should be clear to one skilled in the art, the aspect ratio (length to width) can be altered and achieve similar results.

0080In an alternative example, a piezoresistive transduction scheme can be utilized to measure changes in the position of the deflectable region in the pressure cavity. One or more piezoresistive elements translate mechanical strain into changes in electrical resistance. The piezoresistor is made of, e.g., polysilicon and formed on the interior of the pressure cavity in lieu of the electrodes in previous examples. The resistance modulation is, e.g., detected through a fully active Wheatstone bridge, as is known in the art. Optimally, the Wheatstone bridge configuration used is one where only one leg of the bridge is fixed to the deflectable region of the pressure cavity. This design reduces the number of feedthroughs to two.

0081The lower wall <b>510</b> of the pressure cavity <b>520</b> comprises passages <b>590</b>, <b>595</b> that traverse the sensor body <b>505</b> and are in contact with the electrodes <b>555</b>, <b>560</b>. As shown in <figref idref="DRAWINGS">FIG. 29</figref>, electrical contact pads <b>600</b>, <b>605</b> can be formed on the back side of the electrodes <b>555</b>, <b>560</b> and extend to the exterior of the housing, thereby providing a region on the exterior of the sensor <b>500</b> configured with sufficient dimensions so as to allow for a means for connection with external electronics. As an alternative, the passages <b>590</b>, <b>595</b> can be filled with an electrically conductive material, with contact pads <b>600</b>, <b>605</b> in electrical communication with the electrodes <b>555</b>, <b>560</b> by way of the conductive material <b>610</b>. The electrode <b>555</b>, the passage <b>590</b>, and, if present, the electrical contact pad <b>600</b> and any electrically conductive material <b>610</b> filling the passage <b>590</b> comprises a first electrical feedthrough <b>615</b>. The electrode <b>560</b>, the passage <b>595</b>, and, if present, the electrical contact pad <b>605</b> and any electrically conductive material <b>610</b> filling the passage <b>595</b> comprises a second electrical feedthrough <b>620</b>.

0082One transduction scheme capable of measuring changes in the position of the deflectable region in the pressure cavity is illustrated in <figref idref="DRAWINGS">FIG. 27</figref>. Sensor <b>300</b> and ASIC <b>310</b> together comprise an active Wheatstone bridge, which is known in the art for measuring an unknown resistance. Sensor <b>300</b> comprises a piezoresistor of resistance value R<b>1</b>. Piezoresistors are well known in the art. The other three legs of the Wheatstone bridge comprise resistors <b>312</b>, <b>314</b>, <b>316</b> with values R<b>2</b>, R<b>3</b>, R<b>4</b> respectively. Voltage <b>320</b> of value V<b>0</b> is supplied by a battery (not shown). The circuit operates on the following principle, which discussion is presented for illustrative purposes only. When voltage <b>320</b> is applied with value V<b>0</b>, and R<b>1</b>, R<b>2</b>, R<b>3</b> and R<b>4</b> are all of known values, then the value VS of voltage <b>322</b> may be determined as is well known in the art from knowledge of V<b>0</b>, R<b>1</b>, R<b>2</b>, R<b>3</b>, and R<b>4</b>. However, if the resistance R<b>1</b> of sensor <b>300</b> changes while values R<b>2</b>, R<b>3</b>, and R<b>4</b> of resistors <b>312</b>, <b>314</b>, <b>316</b> remain unchanged, then the value VS of voltage <b>322</b> will change. As is well known in the art, measurement of the changed value VS of voltage <b>322</b> may then be used to determine the value of resistance R<b>1</b> of the sensor <b>300</b>. Because sensor <b>300</b> comprises a piezoresistor, the value R<b>1</b> of sensor <b>300</b> changes in response to a change in position of the deflectable region in the pressure cavity, and this circuit therefore gives a measurement of that change in position.

0083As previously indicated, various capacitor configurations are possible. <figref idref="DRAWINGS">FIG. 28</figref> illustrates a sensor <b>350</b> that includes a pressure cavity body <b>351</b> defining an internal pressure chamber <b>352</b>. One of the walls defining the pressure cavity <b>352</b> comprises a deflectable region <b>354</b> configured to deflect under a physiologically relevant range of pressure. In a preferred embodiment, a wall of the pressure cavity <b>351</b> is thinned relative to other walls of the pressure cavity body to form the deflectable region <b>354</b>.

0084A capacitor comprises a single lower electrode <b>356</b> located on a first wall <b>358</b> of the chamber <b>352</b>. A second electrode <b>360</b> is disposed on an opposite wall <b>362</b> of the pressure cavity <b>352</b> in parallel, spaced apart relation to the lower electrode <b>356</b>. The upper electrode <b>360</b> is mechanically coupled to the deflectable region <b>354</b>.

0085The lower portion of the pressure cavity <b>352</b> contains a pair of passages <b>364</b>, <b>365</b> that traverse the hermetic pressure cavity body <b>351</b>. The first passage <b>364</b> is in contact with the lower electrode <b>356</b>. The second passage <b>365</b> is in contact with the upper electrode <b>360</b> by way of an electrode in the form of an electrically conductive post <b>357</b> disposed within the pressure cavity <b>352</b>. Electrical contact pads <b>366</b>, <b>367</b> are formed within the passages <b>364</b>, <b>365</b> on the back side of the electrodes <b>356</b>, <b>357</b> and extend to the exterior of the housing <b>351</b>, thereby providing a region on the exterior of the sensor <b>350</b> configured with sufficient dimensions so as to allow for a means for connection with external electronics.

0086While the invention as been illustrated in the context of a biological device, it will be appreciated that the hermetic chamber herein described can be adapted to non-biological applications, for example, industrial applications in which a harsh environment is encountered.

0087Specific embodiments have been described herein, by way of example and for clarity of understanding, and variations and modifications to the present invention may be possible given the disclosure above. Hence the scope of the present invention is limited solely by the appended claims.